Fax: (011) 81-99-265-7426
Utility of the GeneSearch breast lymph node assay for the rapid evaluation of sentinel lymph nodes in breast cancer
Version of Record online: 19 AUG 2010
Copyright © 2010 American Cancer Society
Volume 116, Issue 19, pages 4450–4455, 1 October 2010
How to Cite
Funasako, Y., Uenosono, Y., Hirata, M., Arigami, T., Yanagita, S., Arima, H., Ehi, K., Kijima, Y., Yoshinaka, H. and Natsugoe, S. (2010), Utility of the GeneSearch breast lymph node assay for the rapid evaluation of sentinel lymph nodes in breast cancer. Cancer, 116: 4450–4455. doi: 10.1002/cncr.25479
- Issue online: 22 JUN 2010
- Version of Record online: 19 AUG 2010
- Manuscript Accepted: 18 MAY 2010
- Manuscript Revised: 19 APR 2010
- Manuscript Received: 14 JAN 2010
- breast cancer;
- sentinel lymph node;
- breast lymph node assay;
- reverse transcriptase-polymerase chain reaction;
The potential for reducing the need for second surgery for axillary lymph node dissection (ALND) has made the intraoperative evaluation of sentinel lymph nodes (SLNs) attractive. The goal of the current study was to evaluate the clinical application of the breast lymph node (BLN) assay, a real-time reverse transcriptase-polymerase chain reaction assay for SLN metastases, by comparing this test with routine pathologic examination.
A total of 117 patients with breast cancer underwent breast surgery with SLN biopsy. Each SLN was cut in half along the plane of the longest dimension. Half of each lymph node was examined by the 2 markers of the BLN assay, mammaglobin and cytokeratin 19, and the other half was examined by hematoxylin and eosin staining (H&E) and immunohistochemical staining (IHC) for pancytokeratins.
A total of 204 SLNs were obtained from 117 patients. H&E staining identified metastases in 31 SLNs (15.2%), and IHC staining detected metastases in 6 SLNs; 40 SLNs from 32 patients were found to be positive for metastasis using the BLN assay. The assay results were correlated with the pathologic diagnoses by H&E and IHC staining (P <.001). The sensitivity of the BLN assay compared with pathologic findings classified according to the TNM classification was 95.7% for macrometastases, 60.0% for micrometastases, and 55.6% for isolated tumor cells.
The 2-marker BLN assay performs in a manner that is comparable to, and analyzes more tissue than, routine pathologic examination. Therefore, clinical intraoperative use of the BLN assay for SLNs may result in a reduction in the need for second surgery for ALND. Cancer 2010. © 2010 American Cancer Society.
Axillary lymph node (ALN) metastasis is 1 of the most important prognostic factors in breast cancer. The initial standard of care was full ALN dissection (ALND), but when sentinel LN (SLN) biopsy was shown to accurately predict the involvement of the remaining axilla (allowing for a reduction in the morbidity associated with full ALND), it quickly became the standard clinical practice. Today, SLN biopsy is commonly performed to aid in the clinical staging of patients with breast cancer.1, 2 SLNs are typically assessed postoperatively with hematoxylin and eosin (H&E) staining and occasionally with immunohistochemical (IHC) staining. However, these postoperative assessments take several days to provide results. In contrast, intraoperative SLN assessment eliminates the need for patients with positive results to return later to surgery for completion ALND, thereby avoiding additional patient stress and reducing the institution's financial burden. The intraoperative techniques most commonly used to assess SLNs are histologic examination of frozen sections and cytologic examination of touch imprints.3, 4 Both techniques have high specificity; however, they lack the sensitivity of postoperative H&E evaluation, which is considered to provide the definitive diagnosis of lymph node metastases. Although the most likely causes of this lower sensitivity are limited LN sampling and interpretation difficulties, other factors such as interpathologist variability, hard-to-detect cancer types, and nonstandardized procedures also negatively influence performance.5 To improve the accuracy of the intraoperative assessment of SLN status, the entire SLN must be examined in a standardized manner, which is a technique that normally is not feasible in common practice. However, studies have found that molecular approaches, such as real-time reverse transcriptase-polymerase chain reaction (RT-PCR), based on the identification of certain mRNA markers, are useful for the detection of LN metastasis.6, 7 Recently, dedicated kits and instruments for a real-time RT-PCR assay (GeneSearch Breast Lymph Node [BLN] Assay; Veridex LLC, Raritan, NJ) have become commercially available in the United States and Europe for the detection of SLN metastases in the SLNs of patients with breast cancer.8 The BLN assay includes ready-to-use reagents, as well as positive and negative controls. The assay can be performed in a short time (approximately 40 minutes), making it suitable for intraoperative use. In this study, we investigated the clinical application of the BLN assay by comparing the assay with routine histopathologic examination (H&E and IHC) in analyzing SLN status.
MATERIALS AND METHODS
Informed consent was obtained from all patients, in accordance with the ethical standards of the Committee on Human Experimentation of Kagoshima University Hospital, Kagoshima, Japan. Between March 2005 and December 2008, 117 female patients with breast cancer underwent surgical treatment with SLN biopsy at the Kagoshima University Hospital. The patients ranged in age from 23 to 90 years (mean age, 55.4 years). No patient received surgical treatment, chemotherapy, or radiotherapy before SLN biopsy. Patient characteristics are shown in Table 1.
|Clinical T classification|
|Clinical N classification|
Identification of SLN
In this study, technetium-99m-tin colloid measuring 100 nanometers (nm) in size was used as a tracer.9 One day before surgery, approximately 3 millicuries (mCi) (1.5 mL) of the colloid (1 mM of anhydrous stannous chloride solution; Nihon Medi-Physics Company Ltd, Nishinomiya, Japan) were injected into 3 points (0.5 mL each) around the tumor: the nipple, axilla side of the tumor edge, and subcutaneous tissue just above the tumor (Fig. 1). After injection, the area around the injection point was massaged for 5 minutes. Radioisotope injections were administered by 3 physicians (Y.F., K.E., and M.H.). Lymphoscintigraphy was performed using the tracer technique, and planar images (E-Cam Nuclear Gamma Camera; Toshiba, Tokyo, Japan) from the anterior and right and left lateral views were obtained 2 hours after injection of the radioisotope (Fig. 2). During surgery, SLNs were identified using Navigator GPS (Tyco Healthcare, Ltd, Tokyo, Japan). To avoid contamination of the LNs during tumor manipulation, SLNs were excised surgically before breast resection. After SLN removal, the absence of residual radioactivity in the axilla was confirmed intraoperatively using Navigator GPS. When counts for individual LNs were 10 times greater than background levels, the LN was identified as an SLN.
Assessment of LN Metastasis
Each SLN was bisected in the plane of largest dimension. Half of each lymph node was then frozen in liquid nitrogen and stored at −80°C for the BLN assay; the other half was examined intraoperatively by frozen section. The remaining frozen tissue was fixed in 10% formaldehyde and embedded in a paraffin block for postoperative examination (Fig. 3). The paraffin blocks were cut into 3-μm sections and examined by H&E and IHC staining. IHC staining involved the use of a monoclonal anticytokeratin (anti-CK) antibody cocktail (AE1/AE3; Dako Corporation, Carpinteria, CA) that reacts with a broad spectrum of human CKs. IHC sections were deparaffinized in xylene, rehydrated with a graded series of ethanol, and then incubated for 5 minutes in 3% hydrogen peroxide in methanol to block endogenous peroxidase activity. The sections were subsequently immersed in proteinase K (Dako Corporation) to activate the antigen and then incubated with CK monoclonal antibody diluted 1:200 for 30 minutes. After 2 washes of 5 minutes each with phosphate-buffered saline, avidin-biotin complex and immunoperoxidase were applied (ABC method; Vectastain ABC Kit; Vector Laboratories Inc, Burlingame, CA). Cells positive for CK were observed using diaminobenzidine tetrahydrochloride, and the sections were lightly counterstained with hematoxylin. Negative controls were comprised of sections processed in the same manner but without the primary antibody. Primary tumor specimens were used as positive controls. LN metastases were categorized according to the TNM classification as macrometastasis (MA) (those measuring >2 mm), micrometastasis (MI) (those measuring ≤2 mm and >0.2 mm), and isolated tumor cells (ITCs) (those measuring ≤0.2 mm).
GeneSearch BLN Assay
The entire remaining half of the SLN was subjected to the BLN assay performed according to the specifications of the manufacturer. The BLN assay detects the expression of 2 genes: mammaglobin (MG) and cytokeratin 19 (CK19). MG is a marker expressed at high levels in breast tissue and CK19 is an epithelial marker. Both markers are typically not expressed in normal LN tissue, although CK19 may be expressed at low levels. At high levels, both markers are reported to be correlated with the presence of breast cancer metastases.10 The level of expression in the sample is measured by cycle time values. Cycle time value results are applied against predetermined cutoff criteria to provide a qualitative (positive or negative) result. The predetermined cutoff values are established such that the BLN assay is positive only when a metastatic lesion measures >0.2 mm. Furthermore, only 1 of the 2 markers has to be positive for the assay to be considered positive. By using the GeneSearch RNA Sample Preparation Kit (Veridex LLC), the LN tissue is homogenized and placed into RNase-free tubes. The RNA from the tissue homogenate is purified and RT-PCR is performed using the GeneSearch BLN Test Kit (Veridex LLC) and the Cepheid SmartCycler (Cepheid, Sunnyvale, CA) system. The real-time RT-PCR reaction is performed in a fully contained reaction. Three gene markers (MG, CK19, and an internal control of porphobilinogen deaminase) are included in each reaction, and each run contains a positive and negative external control.
Statistical analysis was performed using the chi-square test. A value of P <.05 was considered statistically significant.
SLNs were identified in all patients. A total of 204 SLNs were detected in 117 patients; the mean number of SLNs per patient was 1.73 (range, 1-5 SLNs). H&E staining identified 31 SLNs (15.2%) in 26 patients (22.2%) with SLN metastasis. MA were found in 23 SLNs (18 patients), MI in 5 SLNs (5 patients), and ITCs in 3 SLNs (3 patients). A total of 173 SLNs without metastasis detected by H&E staining were stained by IHC. Six of these SLNs were found to harbor metastases, all of which measured <200 μm and thus were considered to be ITCs.
Each half of the 204 SLNs was examined by the BLN assay markers MG and CK19. There were no failures in internal and external controls. The BLN assay was positive for 40 SLNs from 32 patients; a positive result for both markers was observed in only 17 SLNs (42.5%) from 15 patients. Sixteen SLNs from 12 patients were positive for CK19 only, and 7 SLNs from 5 patients were positive for MG only (Table 2). The lymph node sensitivity of the BLN assay was 83.9% (26 of 31 SLNs) compared with SLNs that were positive on H&E staining only and 66.7% (4 of 6 SLNs) compared with SLNs that were positive on IHC staining only. The results of the assay were found to correlate well with the pathologic diagnosis made on the basis of H&E and IHC staining (P <.001 for each). Ten SLNs from 10 patients were positive on the BLN assay only, with no pathologic evidence of metastatic involvement, even after IHC staining (Table 3). Nine of 10 patients had invasive ductal carcinoma, and pathologic findings of a main tumor in 3 patients demonstrated clear lymphatic vessel invasion. The total number of SLNs with pathologic metastasis and/or BLN assay positivity was 47, and the sensitivity of each examination was 66.0% (31 of 47 SLNs) using H&E staining, 78.7% (37 of 47 SLNs) using IHC staining, and 85.1% (40 of 47 SLNs) using the BLN assay. Metastatic foci of false-negative results using the BLN assay were MAs in 1 SLN, MIs in 2 SLNs, and ITCs in 4 SLNs. When the pathologic findings classified according to the TNM classification were compared with the results of BLN assay, the sensitivity of the assay was 95.7% (22 of 23 SLNs) in MAs, 60.0% (3 of 5 SLNs) in MIs, and 55.6% (5 of 9 SLNs) in ITCs (Table 4).
|CK||MG||No. of LN (N = 204)||No. of Cases (N = 117)|
|+||+||17 (8.3%)||15 (12.8%)|
|+||-||16 (7.8%)||12 (10.3%)|
|-||+||7 (3.4%)||5 (4.3%)|
|-||-||164 (80.4%)||85 (72.6%)|
|BLN assay||40 (19.6%)||32 (27.4%)|
|BLN Assay||H&E (+)||H&E (-)|
|IHC (+)||IHC (-)|
|Positive rate of BLN assay||95.7%||60.0%||55.6%||6.0%|
Data from the current study suggest that the BLN assay has a high performance compared with H&E staining. In addition, its sensitivity is much higher than that of frozen section evaluation, which has been reported to range from 44% to 74% across various studies.3, 11-15 Moreover, the sensitivity of the BLN assay is typically higher than that reported for touch preparations (63% vs 96%).4, 16-18 To the best of our knowledge, the specificity of all available intraoperative tests is comparable. The combination of comparable specificity and the increased intraoperative sensitivity afforded by the BLN assay can potentially prevent second surgery for completion ALND. Moreover, the improvement in the accuracy of an intraoperative diagnosis for LN metastasis may ultimately lead to a better prognosis for patients.
The performance of the BLN assay may be underestimated because of sampling error, especially in the case of smaller metastases. First, the BLN assay and H&E staining examine different sections of LN tissue. In addition, H&E staining is typically limited to a small area of the LN tissue in question. Ideally, a diagnosis of LN metastasis should be based on examination of the total volume of the LN. The BLN assay assesses a relatively large percentage (50%) of the total volume of the LN, but additional metastases could nevertheless be present in the unexamined section. In a study that assessed the accuracy of the BLN assay in detecting SLN metastases in patients with breast cancer, Viale et al found that the assay was positive in 6% of SLNs in which metastases were not detected by standard histologic diagnosis because the metastases were missed as a result of the less-extensive sampling.19 The current study examined half of the SLN volume and evaluated MAs, MIs, and ITCs based on the TNM classification. However, in other reports, the BLN assay examined serial sections of SLNs and evaluated metastases measuring >0.2 mm.5, 20 The sensitivities of MAs and MIs in other reports were similar to those in the current study using half of the SLNs, and ITCs were detected in approximately 55.6% using the BLN assay.5, 19-21 Ten SLNs were assessed as positive on BLN assay only in the current study. Based on the pathologic findings of the main tumor, these patients were determined to have the possibility of a LN metastasis. It was evaluated not as a false-positive finding but as true positive, because histologic examination assessed only the central maximum area and the BLN assay assessed the large volume. Therefore, it is necessary to assess the entire volume of the SLN to achieve a highly sensitive diagnosis using the BLN assay.
There is an additional platform for evaluating the SLNs of patients with breast cancer-the 1-step nucleic acid amplification (OSNA) assay. This assay is characterized by mRNA quantification and requires approximately 21 minutes to complete. OSNA uses CK19 as a marker and reportedly generates results similar to those obtained by histologic staining in 98.2% of LNs from breast cancer patients.22 The GeneSearch BLN assay uses the SmartCycler system. Moreover, the SmartCycler system requires <40 minutes to complete.
In addition to improved sensitivity, the GeneSearch BLN assay offers the advantage of performing a multiple, quantitative RT-PCR assay while simultaneously running an internal control. The availability of internal as well as external controls for every run ensures the quality of the sample and the result. Studies have shown that the use of multiple markers can reduce the risk of obtaining false-negative results, which occurs more commonly with single-marker evaluation. In the current study, the sensitivity of the BLN assay with double markers was 81.1% but declined to 73.0% (27 of 37 SLNs) when only CK19 was considered for a positive result and 51.4% (19 of 37 SLNs) when only MG was considered for a positive result.
The data from the current study indicate that the BLN assay with the double markers CK19 and MG can be a useful tool for the intraoperative assessment of SLNs in patients with breast cancer. The BLN assay performs similarly to H&E staining and, given its ability to evaluate more LN tissue than is possible with standard histologic assessments alone (especially other intraoperative assessments), may allow for a more accurate diagnosis of metastases in SLNs.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 19Comparative evaluation of an extensive histopathologic examination and a real-time reverse-transcription-polymerase chain reaction assay for mammaglobin and cytokeratin 19 on axillary sentinel lymph nodes of breast carcinoma patients. Ann Surg. 2008; 247: 136-142., , , et al.